Jaws of Death Basket

[spyderman]

I came across a thread on FF regarding a "set it and forget it" strategy put forth by Jaws810.
http://www.forexfactory.com/showthread.php?t=379158

It's a cost-averaging system that several of us have been playing with to see if we can get it to trade profitably. It's already changed dramatically from it's original form.

As it seems an easy (easy for me to say anyway) system to code I thought I'd throw it up over here and see if anyone would like to put it together. That would enable us to do some accurate backtesting, hopefully on tradestation (Gary...you out there. )

So here's how I'm trading it on a live account:

Pair: Eur/Usd
I'm choosing this pair as a starting point as it has a good daily average which is beneficial the way I'm trading this.

TF: Irrelevant


So at the beginning of each day (or once a day whenever the EA is set to) the EA fires off a buy and sell order with no s/l and no t/p (obviously hedged). Same thing every day at the same time.

Based on that entry price which is good for the next 24 hours (until the new daily trades are fired) orders are opened when price moves 50 pips. If price moves up 50 a new buy is opened, down a new sell is opened. If price moves another 50 that same day, same thing. A new buy or sell.

If price has been heading up during the day and buys are placed and then it retraces 50 pips no sell order is placed unless it retraces clear down to 50 pips below the initial daily trade.

So in other words the price when the initial daily trades are opened is the baseline for the day. All pending orders are based in 50 pip increments from that point. Baseline resets each day.

As time goes on the daily trades will obviously be hedged against each other. The road to profitability lies in the extra trades that may be placed in the direction of the daily movements. If we get a 150 pip move on a day then we will have 3 extra buy orders in place.

A trend in the long direction, as an example, should result in a nice exponential growth of the stacked buy positions offset by a good sized dd on the sells. So I imagine treating this as a basket system.

Money Management: This as always is a work in progress. As far as an EA goes I'd like the trades to stay open until the trader closes it. Would also be nice to have an option where the EA would close the basket at X pips profit.

Margin: Time (or backtesting) will tell if this may become a problem. Some brokers allow buy and sell orders to offset for margin purposes so would be a much lesser problem.

So that's the short...and long of it

Fire away.

[slipshod]

Jaws of Death, now there's a catchy name One question - are there any rules at all for exit or is it just when the trader feels like doing so? Its pretty important as good exits could make the difference between this making a profit versus a terrible loss.

[spyderman]

Jaws of Death, now there's a catchy name One question - are there any rules at all for exit or is it just when the trader feels like doing so? Its pretty important as good exits could make the difference between this making a profit versus a terrible loss.

I agree. So I think any EA should have an option for closing the whole 'basket' at a user defined amount of pips profit. That would also enable us to optimize backtesting to see what might work best.

At first I was trailing the interday orders but now I think it would be better to leave all trades open until closed.

[spyderman]

Jaws of Death, now there's a catchy name

I was going to use "Jaws of Life" but...time will tell.

[spyderman]

Jaws of Death, now there's a catchy name are there any rules at all for exit

Another option worthy of exploring is trailing all the positions by a percentage of profit after they hit a watermark like 50 pips

[garyfritz]

I think I need a picture to understand exactly how you'd trade it. E.g. you open another buy when it goes up to +50, another one when it gets to +100. Do you open a sell when it drops back to +50, or not until it drops to -50?

This sounds like a system that will make a bundle in clean trending days, and lose its shirt when the market goes back & forth. I'd be pretty surprised if it comes out significantly ahead in the long run.

I can't test hedging in Tradestation. It doesn't do hedging. But you don't need hedging to make this work -- just net out the position. So e.g. at the start of the day you're flat (one long, one short). When it moves up 50 you take your first long, etc.

[Zypip]

Jaws of Death, now there's a catchy name One question - are there any rules at all for exit or is it just when the trader feels like doing so? Its pretty important as good exits could make the difference between this making a profit versus a terrible loss.

I agree. So I think any EA should have an option for closing the whole 'basket' at a user defined amount of pips profit. That would also enable us to optimize backtesting to see what might work best.

Yes, this could be dealt with easily using an instance of MPBM.

Zyp

[spyderman]

Do you open a sell when it drops back to +50, or not until it drops to -50?

-50. I set the baseline price at the daily open (5:00 EDT for me) so all buys and sells for the day will be in 50 pip increments from that point. They would not move with the price movement for the day.

I can't test hedging in Tradestation. It doesn't do hedging. But you don't need hedging to make this work -- just net out the position. So e.g. at the start of the day you're flat (one long, one short). When it moves up 50 you take your first long, etc.

Yeah as I think about it there may be no point in placing the buy and sell at the beginning of the day as they would always be hedged. At least not if we use the basket option and leave all trades open until the whole thing is closed. If however, we decide to trail the orders at a percentage then the daily orders may have more of an impact.

[spyderman]

Jaws of Death, now there's a catchy name One question - are there any rules at all for exit or is it just when the trader feels like doing so? Its pretty important as good exits could make the difference between this making a profit versus a terrible loss.

I agree. So I think any EA should have an option for closing the whole 'basket' at a user defined amount of pips profit. That would also enable us to optimize backtesting to see what might work best.

Yes, this could be dealt with easily using an instance of MPBM.

Zyp

Yeah but can you backtest with MPTM. Might be nice to have the option within the original EA

[garyfritz]

-50. I set the baseline price at the daily open (5:00 EDT for me) so all buys and sells for the day will be in 50 pip increments from that point. They would not move with the price movement for the day.

So basically you have buys at +50, +100, +150, etc, and sells at -50, -100, -150 etc?

If the market trends in one direction, you're golden. If it thrashes around, not so much. You don't have a real clear specification of exit strategy or SL. I assume you don't keep positions open until you hit the opposite side? So e.g. let's say it goes to +101 so now you're long 2. Then it reverses to -101. When do you exit the long positions? With a trailing or BE stop? Presumably that wouldn't have kicked in for the second long which only got +1 open equity.

Still a lot to be clarified here...